Over 90 percent of people under age 65 who have private health insurance obtain it through their employers. In fact, private employers are the largest purchasers of health plan membership. Unless plan payments adequately reflect the expected health needs of a plan's enrollees, health plans have incentives to "risk select," that is, avoid enrollees who are expected to have high costs for health care. The plan may choose not to contract with an employer if the premiums don't adequately reflect the anticipated health care costs of the employees. For this reason, health economists recommend that private employers use a formula based on risk adjusters—such as employee age, sex, health indicators, and prior health care use—to calculate premiums to pay plans, so-called formal risk adjustment. Indeed, Medicare makes extensive use of formal risk adjustment in setting plan premiums.
The conference, "Private Employers and Risk Adjustment," was held in February 2000 to determine why private employers, unlike Medicare, usually do not use formal risk adjustment. The conference was sponsored by the Agency for Healthcare Research and Quality and the Management Sciences Group of the Department of Veterans Affairs. Summaries of a conference overview and six conference papers, supported in whole or part by AHRQ (HS10077), follow.
Glazer, J., and McGuire, T.G. (2001, fall). "Why don't private employers use risk adjustment? Conference overview." Inquiry 38, pp. 242-244.
As the authors of this conference overview point out, employers have mechanisms other than formal risk adjustment to deal with risk differences and adverse selection. These include: introducing regulations into their contracts with health plans; negotiating with plans on the basis of experience (the prior year's health plan expenditures) to set the level of individual and family premiums; paying some part of the costs incurred by plans; and setting employee contributions to accommodate selection concerns (for example, greater contributions for a generally sicker employee population).
Keenan, P.S., Buntin, M.J., McGuire, T.G., and Newhouse, J.P. (2001, fall). "The prevalence of formal risk adjustment in health plan purchasing." Inquiry 38, pp. 245-259.
These researchers examined the frequency with which Medicare, Medicaid, State governments, and private payers made use of formal risk adjustment in their plan payments in 1998. In that year, formal risk adjustment was used for about one-fifth of all enrollees in capitated health plans. Although Medicare and Medicaid relied on formal risk adjustment for virtually all their health plan enrollees, the practice was used for only about 1 percent of the 63.1 million people for whom private employers made health insurance payments. When the 43 percent of insured employees who had a choice of plans and the 23 percent who had a choice of carrier were considered, the use of formal risk adjustment by private employers increased to 3 percent and 5 percent, respectively. Medicare paid by a formula for all of its 5.3 million beneficiaries in health plans, and some risk adjustment (largely on the basis of age and sex) was used for 96 percent of the 9.2 million Medicaid-eligible people in health plans. These findings raise the question of why the public sector has taken one direction in plan payment, and the private sector has taken another.
Glazer, J., and McGuire, T.G. (2001, fall). "Private employers don't need formal risk adjustment." Inquiry 38, pp. 260-269.
Private employers who are purchasing health care coverage for their employees can pick and choose among plans as they please. They have tools that are better than formal risk adjustment for dealing with adverse selection, suggest these authors. For example, one reason for a formal risk adjustment system is to avoid problems of individual access—that is, the inclination of health plans to take only enrollees with low health care costs (for example, young over old employees). However, outright denial of membership to an employee is a violation of a plan's contract with an employer. On the other hand, employers do face group access problems—for example, if one of three plans offered by an employer anticipates getting the sicker (and more costly) group of employees. In this case, employers can pay a higher premium to plans anticipating more costly employees.
Encinosa, W.E., and Selden, T.M. (2001, fall). "Designing employer health benefits for a heterogeneous workforce: Risk adjustment and its alternatives." Inquiry 38, pp. 270-279.
With formal risk adjustment, employers offer insurers larger premium contributions for high-risk workers than for low-risk workers. Although employers rarely use formal risk adjustment, they do use strategies that accomplish some of the same objectives at lower administrative cost, according to these AHRQ researchers. They may link their contributions to health plan characteristics and premiums rather than individual worker risk types—that is, make larger premium contributions to plans with higher premiums, essentially subsidizing employee choice at a fixed percentage rate. Or, employers can set their contributions based on actuarial assessments of plan generosity. This strategy requires actuarial tables to form an estimate of the coverage cost difference between two plans for, say, a high-cost enrollee. The result is an unequal contribution strategy that is essentially identical to a formal risk adjustment strategy. Also, firms can self-fund coverage or offer only a single health plan with a partial employee contribution, which would help them to provide at least some workers with coverage without imposing undue burdens on low-risk workers. The percentage of all plans that required partial or full employee contributions grew steadily from 54 percent in 1979 to 71 percent in 1994.
Reprints (AHRQ Publication No. 02-R029) are available from the AHRQ Publications Clearinghouse.
Feldman, R., Dowd, B.E., and Maciejewski, M. (2001, fall). "A demand-side view of risk adjustment." Inquiry 38, pp. 280-289.
Employers who offer multiple health plans often make larger contributions to the premiums of high-cost plans because without such premium subsidies (a form of demand-side risk adjustment), too few employees would choose the high-cost plans preferred by workers at high risk for health problems, suggest these authors. In fact, their study found that indeed employers were more likely to provide a subsidy when they offered plans preferred by high-risk workers. They estimated a model of the employer premium subsidy, using data from a telephone survey of large public employers in 1994.
Only 47 employers made a level-dollar contribution to single-coverage health plans; in contrast, 216 paid at least part of the extra cost of high-cost plans. Fifty-seven employers paid a level-dollar premium for family plans in contrast to 177 cases with a subsidy for high-cost plans. High-cost plans typically provide easy access to medical specialists, one indicator of health care quality. By reducing an employee's out-of-pocket premium for high-cost plans, an employer can provide an indirect incentive for high-quality care.
Frank, R.G., and Rosenthal, M.B. (2001, fall). "Health plans and selection: Formal risk adjustment vs. market design and contracts." Inquiry 38, pp. 290-298.
According to these authors, three factors reduce the value of risk adjustment from a plan's point of view. First, only a relatively small segment of privately insured Americans face a choice of competing health plans. Second, health plans share much of their insurance risk with payers, providers, and reinsurers. Third, the experience rating (previous year's health care costs for a firm's employees) that occurs during the premium negotiation process and management of benefit coverage appear to substitute for risk adjustment. A 1997 national survey of all U.S. employers with 10 or more employees found that 41 percent of firms negotiated with plans over premiums. A substantial share of both self-insured firms and those that purchase insurance engaged in premium negotiations (42 percent and 38 percent, respectively). Overall, it is the average premium that a plan gets for the enrollees that determines its profits. Plans are essentially indifferent to the way average revenue is packaged.
Ellis, R.P. (2001, fall). "Formal risk adjustment by private employers." Inquiry 38, pp. 299-309.
The reason that adoption of formal risk adjustment has been slow is because many agents, including consumers, employers, health plans, and providers, haven't demanded it, says this author. He describes several reasons why this approach has rarely been used by U.S. private employers: lack of data, difficulties of using potentially distorted signals, need for market power, availability of alternative strategies, and the existing historical structure of health care markets. Data problems are paramount. Employers report that claims information is not reliably coded or not available from all competing health plans on a comparable basis. Survey-based information is too expensive, and individual-level cost information is not available from managed care plans under capitation to evaluate and calibrate payments to these plans. In addition, many large U.S. employers tend to have employees who are geographically dispersed, so they lack the geographic concentration that increases market power and facilitates the negotiation of risk-adjusted premiums.
Return to Contents
Proceed to Next Article
540 Gaither Road Rockville, MD 20850